{"slug": "consistency-models-for-fast-mri-reconstruction-using-regularization-by-denoising", "title": "Consistency Models for Fast MRI Reconstruction Using Regularization by Denoising", "summary": "Researchers propose CM-RED, a new MRI reconstruction method that integrates a pretrained consistency model into the regularization by denoising (RED) scheme, achieving high-quality reconstructions on fastMRI knee and brain datasets using only 4 network function evaluations (NFEs). The method outperforms existing diffusion model- and consistency model-based approaches in quantitative metrics and visual fidelity, and shows strong robustness to hyperparameter variations. Source code and pretrained models are publicly available at https://github.com/MerveGulle/CM-RED.", "body_md": "arXiv:2608.20561v1 Announce Type: cross\nAbstract: Diffusion models (DMs) have emerged as powerful generative priors for MRI reconstruction with promising results. Yet DM-based methods require extensive iterative refinement, limiting their practical deployment. Consistency models (CMs) provide a compelling alternative, aiming to map out the diffusion trajectory in a single pass, enabling faster generation. In this work, we propose CM-RED, a novel MRI reconstruction method that integrates a pretrained CM into the regularization by denoising (RED) scheme. Our method builds on accelerated proximal gradient RED (RED-APG), and further incorporates controlled noise injection during the update steps to enhance generative diversity and accelerate convergence. Extensive experiments on the fastMRI knee and brain datasets demonstrate that CM-RED achieves high-quality reconstructions across multiple anatomies, contrast weights, acceleration factors, and undersampling patterns, using only 4 network function evaluations (NFEs). The proposed method consistently outperforms existing DM- and CM-based approaches in both quantitative metrics and visual fidelity, and exhibits strong robustness to hyperparameter variations, highlighting CM-RED as an efficient and effective generative framework for accelerated MRI reconstruction. The source code and pretrained models are publicly available at https://github.com/MerveGulle/CM-RED.", "url": "https://wpnews.pro/news/consistency-models-for-fast-mri-reconstruction-using-regularization-by-denoising", "canonical_source": "https://www.machinebrief.com/news/consistency-models-for-fast-mri-reconstruction-using-regular-pmo6", "published_at": "2026-08-24 04:00:00+00:00", "updated_at": "2026-08-24 06:14:14.915598+00:00", "lang": "en", "topics": ["machine-learning", "generative-ai", "ai-research"], "entities": ["CM-RED", "fastMRI", "arXiv", "MerveGulle"], "alternates": {"html": "https://wpnews.pro/news/consistency-models-for-fast-mri-reconstruction-using-regularization-by-denoising", "markdown": "https://wpnews.pro/news/consistency-models-for-fast-mri-reconstruction-using-regularization-by-denoising.md", "text": "https://wpnews.pro/news/consistency-models-for-fast-mri-reconstruction-using-regularization-by-denoising.txt", "jsonld": "https://wpnews.pro/news/consistency-models-for-fast-mri-reconstruction-using-regularization-by-denoising.jsonld"}}